The experiments that demonstrated an association of the SV40 large T antigen (T) or Adenovirus E1A with the retinoblastoma tumor suppressor gene product (RB) established a molecular basis for DNA tumor virus- mediated transformation. The interaction of T/E1A with RB is thought to relieve the growth suppression properties of RB thereby permitting cellular transformation. Transformation is dependant on a domain of T (residues 102-115) and conserved regions 1 and 2 of E1A. However, mutations in these domains, that render T/E1A transformation defective, not only disrupt RB binding but also disrupt binding to a family of cellular proteins that resemble RB including p130, p107, p90, and p80. Since mutations in T or E1A will disrupt binding to the entire RB family, it has not been possible to discern the specific role of each in T/E1A- mediated transformation. We will determine if wild type T or E1A can transform primary fibroblasts or immortalized derivatives prepared from RB -/- embryos and, if so, whether mutant T/E1A species, defective in binding to the RB family of proteins, can also transform these cells. The recent development of an RB knockout mouse strain allows a critical examination of the role of the other members of the RB family in T/E1A- mediated transformation. The proposed studies are aimed at determining whether the other members of the RB family participate in T/E1A-mediated transformation. We will determine if wild type T or E1A can transform primary fibroblasts or immortalized derivatives prepared from RB -/- embryos and, if so, whether mutant T/E1A species, defective in binding to the RB family of proteins, can also transform these cells. If RB- binding defective mutants of T or E1A cannot transform the RB -/- cells, this would suggest that at least one additional member of the RB family acts as a tumor suppressor. Preliminary data suggests that at least 2 other proteins, p90 and p80, may belong to the RB family based on their ability to bind to the E1A. We propose to clone the cDNAs for p90 and p80. Once the relevant clones have been obtained, we could begin to characterize p90 and p80 to gain insight into how they function and to determine whether they exhibit tumor suppression activity.